Announcements I: January 10, 2008

Special Seminar Series on Radiological Informatics: As part of a special series on radiological informatics, we are offering seminars on Jan. 14th, 23rd, and 28th. Each seminar is at 12 noon in Alway M104 unless otherwise indicated. Please watch future announcements for each seminar's title and abstract. In addition to the three seminars listed below, there will be at least two more talks, which are being scheduled now and will be posted shortly. Please contact Dr. Sandy Napel for more information.

1) Monday, Jan 14th:
James Z. Wang, PhD
Carnegie Mellon University and Pennsylvania State University

Title:
"A Data-Driven Approach Toward Knowledge Discovery and Improving Healthcare"

Abstract:
Radiology and biomedical informatics are revolutionizing healthcare. It has been predicted that a shortage of trained radiologists will continue in the next three decades. Effective computerized tools will therefore be in great demand. Radiology departments today generate an incredibly massive amount of digital medical images and metadata. Conventional PACS search methods allow physicians to locate images using metadata stored in relational databases. Much more can be done to leverage this wealth of data. Using massively parallel computers, we can mine millions of electronic medical records and millions of high-resolution, high-dimensional, multi-spectrum medical images to draw conclusions statistically based on past cases. We need to invent computational methods to harness the breathtaking quantity of digital information effectively and to generate biomedical knowledge at a pace we could not have imagined. In the last decade, my research group attempted to reduce the significant gap between low-level features extracted from images and high-level semantic concepts. Machine learning, statistical modeling, and mathematical tools have been utilized. I will introduce some of our past research results of relevance to the radiology community. Specifically, the talk will cover the SIMPLIcity visual similarity search, the 3-D hidden Markov models for analyzing volume images, the Automatic Linguistic Indexing of Pictures system, and the ontology-based annotation and retrieval of histological images and quantitative phenotypes. In the coming years, I plan to collaborate with radiologists, physicians, and biologists in order to develop indexing, retrieval, and mining algorithms and systems for large amounts of radiological images and patient-specific data.

Biography:
James Z. Wang is currently a visiting professor at the Robotics Institute of Carnegie Mellon University. He is also a tenured faculty member at Pennsylvania State University. He received a summa cum laude bachelor's degree in mathematics and computer science from the University of Minnesota. From Stanford University, Dr. Wang has received an MS in mathematics, an MS in computer science, and a PhD degree in medical information sciences. He has been a recipient of a National Science Foundation (NSF) Career award and the endowed PNC Technologies Career Development Professorship. Research interests of his group include automatic image tagging, semantics-sensitive image retrieval, image security, biomedical informatics, computational aesthetics, story picturing, art image retrieval, and computer vision. The group has published two monographs and more than 20 journal articles. Science media including Discovery News, Scientific American, National Public Radio, and MIT Technology Review, as well as wired news agencies, have reported his research.

2) Wednesday, January 23rd:
Julia Patriarche, PhD
Mayo Clinic

3) Monday, January 28th:
Jianming Liang, PhD
Siemens Medical Solutions

Comments

Post a comment